When a news reporter interviews a subject, the interview is typically recorded with two video cameras—one covering the reporter and one covering the subject—and with a microphone that records both voices. The output of one video camera is stored in one video recording, the output of the second video camera is stored in a second video recording, and the output of the microphone is stored in an audio recording. FIG. 1 depicts a scene of such an interview.
Video camera 102-1 films Jones 102 and its output is stored on video-recorder 103-1. Video camera 102-2 films Smith 101 and its output is stored on video-recorder 103-2. Microphone 105 picks up the voice of both Smith 101 and Jones 102 and its output is recorded on audio recorder 104.
To facilitate the editing of the interview, a written transcript of the interview is produced from the audio recording. FIG. 2 depicts a typical written transcript as it is well known in the prior art. The transcript comprises two columns of text. The second or right-hand column comprises a textual representation of the words spoken by Mr. Smith and the words spoken by Ms. Jones. The text alternates between the words of Mr. Smith and the words of Ms. Jones, and reflects the back-and-forth nature of a typical interview.
The first or left-hand column comprises a textual representation of one or more time codes. A time code is a signal that is generated and distributed by one of the video cameras and is recorded with each audio and video recording.
Each time code is a more-or-less-arbitrarily-chosen reference value that enables the temporal correlation of the video and audio recordings. Typically, the time code is represented in an hour:minute:second.frame format. Despite the name of the code and its hour:minute:second:frame format, the time code is not related to the real time at which the recording was made, and might or might not bear a relation to the elapsed time at which the recording was made.
On the transcript, each time code is associated with a passage of text. For example, as depicted in FIG. 2 the time code “10:23:20.29” is associated with the portion of transcript 201 that is itself associated with the video recording in which Ms. Jones can be heard uttering “Well, you don't want to rush . . . .”
The time codes perform two functions. First, the time codes are printed on the transcript so that the editor of the interview can correlate each passage of text with its position in the audio and video recordings. This helps the editor locate and view the video and audio recordings to ensure that the speaker's mien is consistent with the import of the written transcript. Second, the presence of the time codes in the audio and video recordings enables the editor to perfectly synchronize the video recordings with each other and with the audio recording.
The need exists, however, for a system for editing audio and video recordings that is more advantageous than systems in the prior art.